A full-length cDNA clone was constructed in which the entire Norwalk virus (NV) genome was engineered downstream from the T7 promoter. Analysis of this clone in a coupled in vitro transcription and translation system containing T7 polymerase showed that the polyprotein encoded by ORF1 was proteolytically processed by an active protease and produced three products of the correct predicted size (45, 40, and 114 kDa). The three predicted cleavage products encoded in the NV ORF1 have been engineered individually into the pCI mammalian expression vector that contains a T7 promoter for in vitro transcription and translation and a CMV promoter for expression in mammalian cells. The resulting three translation products produced in the coupled T7 polymerase-based transcription and translation reaction were identical in size to those produced by in vitro proteolytic processing of the full-length clone. The NV cleavage products were compared to those produced from the cultivatable feline calicivirus (FCV) full-length clone. The FCV nonstructural polyprotein is processed into several mature cleavage products in vitro that are similar to those produced in infected cells, suggesting that the FCV-encoded protease functions efficiently in the absence of host cell factors. The difference in the number of ?mature? cleavage products between FCV and the human caliciviruses may suggest that complete proteolytic processing of the human calicivirus polyprotein is inefficient in vitro. It is not yet clear whether incomplete proteolytic processing of the NV nonstructural polyprotein is one of the blocks in achieving human calicivirus replication in cells. The pCI-based plasmids containing NV genes are being used to study proteolytic processing and localization of nonstructural protein expression in mammalian cells. The Sindbis virus expression system has been explored this year as a method to produce biologically active NV recombinant proteins in mammalian cells. Sindbis ?pseudovirions? have been produced that carry various Norwalk virus structural and nonstructural proteins. Our goal is to determine whether functional proteins can be produced in high levels in this system. It may be possible to provide Norwalk virus replication enzymes in ?trans? that will replicate transfected NV RNA molecules. We will also continue our use of the vaccinia virus expression system as an approach to the expression of ?authentic? NV and Hawaii virus structural and nonstructural proteins. Discrete regions of the NV ORF1 that span the entire polyprotein have been cloned into a baculovirus vector that allows the expression of these regions as GST fusion proteins. In addition, several regions of the NV genome have been expressed in the pET bacterial system and, in addition, antisera have been raised in guinea pigs. Thus far, we have developed antibodies specific for the ?3A? region (by analogy with the picornaviruses) and the ?3D? polymerase region. In addition, we have raised antisera in rabbits to synthetic peptides deduced from three predicted antigenic sites of the nonstructural polyprotein. The antibodies raised against the putative viral VPg peptide immunoprecipitated the 114 kDa cleavage product (that contains the predicted VPg region) derived by in vitro translation of the NV ORF1 clone. Future experiments will examine whether this antisera reacts with the native viral VPg associated with genomic RNA. In addition, the ORF3 product of NV has been expressed in bacteria and purified. Antibodies have been raised in guinea pigs that are specific for this protein. Functional studies of the ORF3 product are in progress. This year, a CRADA was established with InCell Corporation in San Antonio, Texas. This company specializes in the in vitro cultivation of human cells. Two human colon cells lines were provided to our laboratory and we are presently ?passaging? NV in roller tube cultures. Because these cells are derived from the human gut, it is possible that they may support viral replication. We will continue to passage NV in the absence of visible cytopathic effect or positive immunofluorescence multiple times in an attempt to select a variant of the virus that replicates in vitro. We also plan to explore other intestinal cells lines provided by InCell as substrates for NV replication.